Fast Monte Carlo Simulation For Estimation Maximum Expected Drawdown (run python code x2000 faster)

In this video we will see how with a few simple steps we can improve execution performance of a python code more than in x1000 time. And in order for us to be not only interesting but also useful, we will perform all our manipulations using the example of Monte Carlo simulation to calculate an expected maximum drawdown.

00:00 Intro
00:26 Maximum Drawdown definition
00:53 Maximum Drawdown calculation example
01:30 Maximum Drawdown python implementation
02:26 Maximum Drawdown python implementation Numba version
03:26 Monte Carlo Method and definition of our goals
04:30 Stock price series generation theory
06:00 Calculation of the price series characteristics
06:22 Implementation of the function for price series generation
07:25 Example of the price series generation
07:45 Implementation of the Monte Carlo simulation for estimation of the expected MDD
08:05 Comparison of the real expected MDD with the calculated from the historical price series
08:45 Monte Carlo Simulation V0 execution timing measurement
09:05 Monte Carlo Simulation V1 Implementation and timing measurement
09:45 Monte Carlo Simulation V2 Implementation and timing measurement
10:32 Convert V0, V1, V2 into the Numba code and timing measurement
11:20 Monte Carlo Simulation V3 Implementation and timing measurement
12:22 Monte Carlo Simulation V3 C++ Implementation and timing measurement
12:39 Monte Carlo Simulation V3 Numba CUDA Implementation and timing measurement
14:07 Monte Carlo Simulation V3 C++ CUDA Implementation and timing measurement
14:30 CuPy definition
14:53 Monte Carlo Simulation V3 CuPy CUDA Implementation and timing measurement
15:37 Results analysis

Code: https://github.com/CloseToAlgoTrading/CodeFromVideo/tree/master/Episode_MonteCarlo_MDD